As of 13 August 2026, AI can predict when you need to reorder stock.

What to actually do

1. Use a tool built for this

   The route this page recommends

2. Do it yourself

   Second choice

   A chat interface, chat-fluent skill, and roughly 1 hour until you can act on the result.

   How to actually do it

   1. Open your stock system, spreadsheet or sales reports and export recent item-level sales, current stock, open purchase orders, pack sizes and minimum order quantities.
   2. Gather each supplier's current lead time, delivery schedule, order cut-off information and any known closures or reliability problems.
   3. Mark promotions, seasonal peaks, product launches, discontinued lines, unusual bulk orders and stockout periods so the model does not treat them as ordinary demand.
   4. Paste the records and the method in the prompt into a chatbot, then ask it to return the reorder table and show every calculation.
   5. Compare the current-stock and sales figures in the answer with your source records, and recalculate a sample of the demand and reorder-point calculations in your spreadsheet.
   6. Check each flagged assumption with the relevant buyer, warehouse lead or supplier, then change the safety-stock policy or exclusions in the prompt and rerun the table.
   7. Approve the final suggested order quantities against your cash, storage capacity and supplier terms, and enter orders into your normal purchasing system yourself.

   Prompt

   Act as an inventory planning analyst. Using only the data I provide, calculate when each stock item should be reordered and produce a table with these columns: item, current stock, average daily demand, demand during supplier lead time, safety stock, reorder point, suggested order date, suggested order quantity, and the reason for the recommendation. Use this method unless the data clearly requires another one: average daily demand equals total units sold divided by the number of days covered; reorder point equals demand during lead time plus safety stock; suggested order date is when projected stock reaches the reorder point; suggested order quantity covers the stated review period while respecting the minimum order quantity and pack size. State every assumption, identify missing or inconsistent data, and do not invent sales, lead times, stock levels, prices or dates. Separate calculated results from assumptions. Flag items affected by seasonality, promotions, supplier unreliability, new products, discontinued products or unusual one-off demand instead of treating them as normal demand. Show the calculation for each recommendation so I can check it. Do not place orders or claim to know current stock unless it appears in the data below. Data: [paste sales history, current stock records, supplier lead times, minimum order quantities, pack sizes, review period, safety-stock policy and relevant business notes here].

   Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.

3. Hand it to a person

   The distant third

   A person who owns the outcome does this end to end, worth it when the failure is dear.